Name Generation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from utils_rnn import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59587 total characters & 27 unique characters.\n"
     ]
    }
   ],
   "source": [
    "# Data cleaning\n",
    "\n",
    "# combined_names.txt contains all the Nepali names\n",
    "data = open('combined_names.txt', 'r').read()\n",
    "data= data.lower()\n",
    "characters = list(set(data))\n",
    "characters = sorted(characters)\n",
    "\n",
    "data_size, vocab_size = len(data), len(characters)\n",
    "print('%d total characters & %d unique characters.' % (data_size, vocab_size))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding indexes -- > characters and reverse\n",
    "char_to_idx = { ch:i for i,ch in enumerate(characters) }\n",
    "idx_to_char = { i:ch for i,ch in enumerate(characters) }"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function for sampling new names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sample(parameters, char_to_idx):\n",
    "    \n",
    "    \n",
    "    Waa, Wax, Wya, by, b = parameters['Waa'], parameters['Wax'], parameters['Wya'], parameters['by'], parameters['b']\n",
    "    vocab_size = by.shape[0]\n",
    "    n_a = Waa.shape[1]\n",
    "    x = np.zeros((len(char_to_idx),1))\n",
    "    a_prev = np.zeros((n_a,1))\n",
    "        # Create an empty list of indices, this is the list which will contain the list of indices of the characters to generate (≈1 line)\n",
    "    indices = []\n",
    "    # idx is the index of the one-hot vector x that is set to 1\n",
    "    # All other positions in x are zero.\n",
    "    # We will initialize idx to -1\n",
    "    idx = -1 \n",
    "    counter = 0\n",
    "    newline_character = char_to_idx['\\n']\n",
    "    while (idx != newline_character and counter != 15) : \n",
    "        a = np.tanh(np.dot(Wax,x) + np.dot(Waa,a_prev) + b)\n",
    "        z = np.dot(Wya,a) + by\n",
    "        y = softmax(z)\n",
    "        # Ssampling the indices from the probability distribution of y\n",
    "        idx = np.random.choice(range(len(y)), p = y.ravel())\n",
    "        indices.append(idx)\n",
    "        \n",
    "        # X[idx] = 1 because X will be passed to the next timestep as Y<t>\n",
    "        x = np.zeros((vocab_size,1))\n",
    "        x[idx] = 1\n",
    "        a_prev = a\n",
    "        if (counter == 15):\n",
    "            indices.append(char_to_idx['\\n'])\n",
    "    \n",
    "    return indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Defining a model\n",
    "def model(data, filename, idx_to_char, char_to_idx, num_iter = 35000, n_a = 50, nepali_name = 10, vocab_size=27):\n",
    "    \n",
    "    n_x, n_y = vocab_size, vocab_size\n",
    "    loss = -np.log(1.0/vocab_size)*nepali_name\n",
    "    loss_points = []\n",
    "    generated_name_list =[]\n",
    "    \n",
    "    parameters = initialize_parameters(n_a, n_x, n_y)\n",
    "    with open(filename) as f:\n",
    "        training_examples = f.readlines()\n",
    "    training_examples = [x.lower().strip() for x in training_examples]\n",
    "    # Initialize the hidden state \n",
    "    a_prev = np.zeros((n_a, 1))\n",
    "    np.random.shuffle(training_examples)\n",
    "    \n",
    "    for j in range(num_iter):\n",
    "        # Mod ie. % returns back to index 0 once we reach end of the examples\n",
    "        idx = j % len(training_examples)\n",
    "        # Get an example from idx \n",
    "        single_training_example = training_examples[idx]\n",
    "        # Get all the characters from the single training example\n",
    "        single_training_character = [char for char in single_training_example]\n",
    "        # Get the indexes of the characters in the single training example\n",
    "        single_example_idx = [char_to_idx[ch] for ch in single_training_character]\n",
    "        # None prepended to set the input vector to 0 vector\n",
    "        X = [None] + single_example_idx\n",
    "        idx_newline = char_to_idx[\"\\n\"]\n",
    "        Y = X[1:] + [idx_newline]\n",
    "        \n",
    "        curr_loss, gradients, a_prev = optimize(X,Y,a_prev,parameters,learning_rate=0.01)\n",
    "        loss = smooth(loss, curr_loss)\n",
    "        \n",
    "        if j % 2000 == 0:\n",
    "            \n",
    "            print('Iteration: %d, Loss: %f' % (j, loss) + '\\n')\n",
    "            loss_points.append(loss)\n",
    "            \n",
    "            for name in range(nepali_name):\n",
    "                \n",
    "                # Sample indices and print them\n",
    "                sampled_indices = sample(parameters, char_to_idx)\n",
    "                generated_name_list.append(print_sample(sampled_indices, idx_to_char))\n",
    "      \n",
    "            print('\\n')\n",
    "        \n",
    "    return loss_points, generated_name_list\n",
    "\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration: 0, Loss: 32.945184\n",
      "\n",
      "Vgpjlvde\n",
      "J\n",
      "Kxzbufnvmnkvjviqiqtvcijqmextpulbajxjkodwgqvhkgbyaysdvjrshnyjeyguezw\n",
      "Lpjsyhhalyj\n",
      "Xa\n",
      "\n",
      "Nrlfscbuqncrteyxjrbacgtghcpp\n",
      "Eurdzg\n",
      "Vqciuddhh\n",
      "Vglbz\n",
      "\n",
      "\n",
      "Iteration: 2000, Loss: 21.674202\n",
      "\n",
      "\n",
      "Dlraessirenant\n",
      "Er\n",
      "Danasujotp\n",
      "\n",
      "Di\n",
      "Taakinsayomanhiban\n",
      "R\n",
      "Tphie\n",
      "Ushasashamircridgshbeqeshan\n",
      "\n",
      "\n",
      "Iteration: 4000, Loss: 18.445728\n",
      "\n",
      "Bamijabhanasamimodishu\n",
      "Raraniyarolakhaldabun\n",
      "Punila\n",
      "Ekt\n",
      "Shibtamunx\n",
      "Ta\n",
      "Fadakshil\n",
      "Ujira\n",
      "Hasha\n",
      "Rominoshhedhirramuska\n",
      "\n",
      "\n",
      "Iteration: 6000, Loss: 17.596099\n",
      "\n",
      "Aridha\n",
      "Arodlenyi\n",
      "Sokovi\n",
      "Emesi\n",
      "R\n",
      "Ribdimsh\n",
      "Obava\n",
      "Shabida\n",
      "Ajerkwni\n",
      "Pyobiro\n",
      "\n",
      "\n",
      "Iteration: 8000, Loss: 17.052987\n",
      "\n",
      "Ibi\n",
      "Pusad\n",
      "Ranakdendra\n",
      "Akgi\n",
      "Sheet\n",
      "Richesha\n",
      "Sageeshsa\n",
      "Bishi\n",
      "Pit\n",
      "Idgushanta\n",
      "\n",
      "\n",
      "Iteration: 10000, Loss: 16.607967\n",
      "\n",
      "Shpa\n",
      "Chandrry\n",
      "Ranvi\n",
      "Busam\n",
      "Sjopraprakhan\n",
      "Tusam\n",
      "Ratandrishina\n",
      "Eoday\n",
      "Sobipena\n",
      "Ramba\n",
      "\n",
      "\n",
      "Iteration: 12000, Loss: 16.239223\n",
      "\n",
      "So\n",
      "Reshyal\n",
      "Aesasha\n",
      "Eel\n",
      "Aadhaa\n",
      "Binine\n",
      "Oek\n",
      "Sabuna\n",
      "Sukarant\n",
      "Bikubhana\n",
      "\n",
      "\n",
      "Iteration: 14000, Loss: 16.081384\n",
      "\n",
      "Sumal\n",
      "Ashem\n",
      "Peeza\n",
      "Bimash\n",
      "Aranak\n",
      "Chadhi\n",
      "Ajtin\n",
      "Bhana\n",
      "Meetshin\n",
      "Anit\n",
      "\n",
      "\n",
      "Iteration: 16000, Loss: 16.193415\n",
      "\n",
      "Chhara\n",
      "Mndhesh\n",
      "Siray\n",
      "Rijbul\n",
      "Gitea\n",
      "Bman\n",
      "Rjish\n",
      "Vaughadipha\n",
      "Barish\n",
      "Ubheche\n",
      "\n",
      "\n",
      "Iteration: 18000, Loss: 16.051097\n",
      "\n",
      "Biyuta\n",
      "Sidwo\n",
      "Bandarkanan\n",
      "Kitru\n",
      "Vinipja\n",
      "Bhilit\n",
      "Ojkazraraksha\n",
      "Sermamana\n",
      "Amit\n",
      "Sargi\n",
      "\n",
      "\n",
      "Iteration: 20000, Loss: 15.858308\n",
      "\n",
      "Pubiqshamaninana\n",
      "Hantulas\n",
      "Avhewan\n",
      "Pman\n",
      "Bejan\n",
      "Bunkra\n",
      "Divanitr\n",
      "Ulidna\n",
      "Anil\n",
      "Ana\n",
      "\n",
      "\n",
      "Iteration: 22000, Loss: 15.770416\n",
      "\n",
      "Bidhendro\n",
      "Anfandura\n",
      "Meedritand\n",
      "Jikrubh\n",
      "Goshwokta\n",
      "Shushskhita\n",
      "Bura\n",
      "Febhan\n",
      "Usweedra\n",
      "Ulur\n",
      "\n",
      "\n",
      "Iteration: 24000, Loss: 15.878279\n",
      "\n",
      "Mitant\n",
      "Aven\n",
      "Ma\n",
      "Ah\n",
      "Sumbig\n",
      "Rimeutesh\n",
      "Shikh\n",
      "Gyuta\n",
      "Binot\n",
      "Vishut\n",
      "\n",
      "\n",
      "Iteration: 26000, Loss: 15.825496\n",
      "\n",
      "Beepash\n",
      "Derij\n",
      "Shnavarash\n",
      "Karushar\n",
      "Utama\n",
      "Dechin\n",
      "Sambara\n",
      "Upchusha\n",
      "Paroh\n",
      "Umbar\n",
      "\n",
      "\n",
      "Iteration: 28000, Loss: 15.665919\n",
      "\n",
      "Lori\n",
      "Riwadi\n",
      "Ilanpa\n",
      "Oniv\n",
      "Jitrar\n",
      "Ayun\n",
      "Sayam\n",
      "Batili\n",
      "Rechit\n",
      "Aamshenj\n",
      "\n",
      "\n",
      "Iteration: 30000, Loss: 15.624969\n",
      "\n",
      "Ak\n",
      "Shojan\n",
      "Karinda\n",
      "Majsh\n",
      "Matrashi\n",
      "Jesan\n",
      "Bajanu\n",
      "Mevya\n",
      "Minima\n",
      "Disji\n",
      "\n",
      "\n",
      "Iteration: 32000, Loss: 15.776632\n",
      "\n",
      "Lopeena\n",
      "Bhanjaman\n",
      "Swisher\n",
      "Mvaj\n",
      "Pkarga\n",
      "Khava\n",
      "Shilbhardri\n",
      "Ishina\n",
      "Jindlanata\n",
      "Puskra\n",
      "\n",
      "\n",
      "Iteration: 34000, Loss: 15.724650\n",
      "\n",
      "Bigesh\n",
      "Yan\n",
      "Lishuk\n",
      "Adhindh\n",
      "Kayar\n",
      "Arawuk\n",
      "Rugan\n",
      "Ashita\n",
      "Roswesh\n",
      "Perisa\n",
      "\n",
      "\n",
      "Iteration: 36000, Loss: 15.533406\n",
      "\n",
      "Rishree\n",
      "Divmanib\n",
      "Sodev\n",
      "Sirsha\n",
      "Chadabdhan\n",
      "Birjanandhi\n",
      "Barendh\n",
      "Jesv\n",
      "Paned\n",
      "Bidesmvam\n",
      "\n",
      "\n",
      "Iteration: 38000, Loss: 15.503872\n",
      "\n",
      "Wadishy\n",
      "Jaku\n",
      "Chalanserena\n",
      "Jughiramal\n",
      "Prawana\n",
      "Risiy\n",
      "Surats\n",
      "Olani\n",
      "Acjhita\n",
      "Burap\n",
      "\n",
      "\n",
      "Iteration: 40000, Loss: 15.704457\n",
      "\n",
      "Unja\n",
      "Rajwoshya\n",
      "Viscmran\n",
      "Kusumarw\n",
      "Hineja\n",
      "Hehan\n",
      "Puput\n",
      "Chanjesh\n",
      "Jopan\n",
      "Mitenoa\n",
      "\n",
      "\n",
      "Iteration: 42000, Loss: 15.659650\n",
      "\n",
      "Upassh\n",
      "Chabish\n",
      "Rarag\n",
      "Aal\n",
      "Bikpral\n",
      "Irlaba\n",
      "Hinimendranamad\n",
      "Angyar\n",
      "Shronj\n",
      "Pili\n",
      "\n",
      "\n",
      "Iteration: 44000, Loss: 15.468905\n",
      "\n",
      "Fhayant\n",
      "Rinva\n",
      "Aevashri\n",
      "Bikunju\n",
      "Rurendra\n",
      "Iyash\n",
      "Anitan\n",
      "Bhanan\n",
      "Bikprita\n",
      "Biman\n",
      "\n",
      "\n",
      "Iteration: 46000, Loss: 15.407870\n",
      "\n",
      "Bilesh\n",
      "Micafadi\n",
      "Bheshu\n",
      "Sanda\n",
      "Emita\n",
      "Ubhira\n",
      "Azaashrolendrana\n",
      "Yahash\n",
      "Chikayu\n",
      "Bilu\n",
      "\n",
      "\n",
      "Iteration: 48000, Loss: 15.614761\n",
      "\n",
      "Jummun\n",
      "Bitur\n",
      "Chmita\n",
      "Jehandra\n",
      "Ankaryan\n",
      "Bimiemabhan\n",
      "Mashri\n",
      "Mhit\n",
      "Binmin\n",
      "Umking\n",
      "\n",
      "\n",
      "Iteration: 50000, Loss: 15.627540\n",
      "\n",
      "Hirdha\n",
      "Sogaj\n",
      "Seplchar\n",
      "Londi\n",
      "Bishna\n",
      "Ewoch\n",
      "Sumsubheti\n",
      "Pudarg\n",
      "Wonisa\n",
      "Bhilan\n",
      "\n",
      "\n",
      "Iteration: 52000, Loss: 15.410767\n",
      "\n",
      "Aniktan\n",
      "Bigaram\n",
      "Bitesh\n",
      "Bind\n",
      "Beesay\n",
      "Ayish\n",
      "Ind\n",
      "Ibalyanta\n",
      "Bualri\n",
      "Alant\n",
      "\n",
      "\n",
      "Iteration: 54000, Loss: 15.378313\n",
      "\n",
      "Rojan\n",
      "Supran\n",
      "Anuna\n",
      "Chikkarsh\n",
      "Bhapa\n",
      "Rishu\n",
      "Beschar\n",
      "Artoz\n",
      "Enika\n",
      "Besmok\n",
      "\n",
      "\n",
      "Iteration: 56000, Loss: 15.603167\n",
      "\n",
      "Reebil\n",
      "Mormo\n",
      "Baripang\n",
      "Dmondra\n",
      "Kitingdha\n",
      "Sharamani\n",
      "Sarudeebe\n",
      "Mibwatshi\n",
      "Bhivik\n",
      "Suwabin\n",
      "\n",
      "\n",
      "Iteration: 58000, Loss: 15.610984\n",
      "\n",
      "Sondra\n",
      "Bigdaperdinish\n",
      "Prakan\n",
      "Sul\n",
      "Aksamta\n",
      "Ashnim\n",
      "Bishas\n",
      "Orsendhaxend\n",
      "Oobakdha\n",
      "Bhukama\n",
      "\n",
      "\n",
      "Iteration: 60000, Loss: 15.365319\n",
      "\n",
      "Yavina\n",
      "Satbia\n",
      "Rguna\n",
      "Alejish\n",
      "Sinj\n",
      "Ipen\n",
      "Shoran\n",
      "Pashtra\n",
      "Abina\n",
      "Yalban\n",
      "\n",
      "\n",
      "Iteration: 62000, Loss: 15.364336\n",
      "\n",
      "Bheth\n",
      "Juna\n",
      "Anikh\n",
      "Renim\n",
      "Syushakeky\n",
      "Asbil\n",
      "Rishu\n",
      "Sotuka\n",
      "Gdrij\n",
      "Jayar\n",
      "\n",
      "\n",
      "Iteration: 64000, Loss: 15.525945\n",
      "\n",
      "Jutur\n",
      "Yyas\n",
      "Alfuj\n",
      "Bhik\n",
      "Pulma\n",
      "Aigu\n",
      "Ulwar\n",
      "Bhupa\n",
      "Rati\n",
      "Bhikrak\n",
      "\n",
      "\n",
      "Iteration: 66000, Loss: 15.607290\n",
      "\n",
      "Rishat\n",
      "Churesh\n",
      "Bishripan\n",
      "Bita\n",
      "Bika\n",
      "Geeshma\n",
      "Hisma\n",
      "Bista\n",
      "Etira\n",
      "Irisu\n",
      "\n",
      "\n",
      "Iteration: 68000, Loss: 15.334564\n",
      "\n",
      "Nindha\n",
      "Uskam\n",
      "Uswal\n",
      "Langar\n",
      "Aulsta\n",
      "Jeetakhkish\n",
      "Bitey\n",
      "Pulsan\n",
      "Eena\n",
      "Eeteri\n",
      "\n",
      "\n",
      "Iteration: 70000, Loss: 15.313010\n",
      "\n",
      "Ishida\n",
      "Krijechar\n",
      "Subi\n",
      "Jighir\n",
      "Biswura\n",
      "Meet\n",
      "Riyetda\n",
      "Aitacaish\n",
      "Mevika\n",
      "Ruphash\n",
      "\n",
      "\n",
      "Iteration: 72000, Loss: 15.550367\n",
      "\n",
      "Iniha\n",
      "Achnab\n",
      "Aalban\n",
      "Lhanu\n",
      "Bhriti\n",
      "New\n",
      "Nafu\n",
      "Uugla\n",
      "Binkuita\n",
      "Udhuurod\n",
      "\n",
      "\n",
      "Iteration: 74000, Loss: 15.614420\n",
      "\n",
      "Urnihhesh\n",
      "Ugyan\n",
      "Runashwar\n",
      "Babish\n",
      "Uktarendra\n",
      "Uprangsu\n",
      "Rablakan\n",
      "Ujerb\n",
      "Ilinishanaa\n",
      "Bigdang\n",
      "\n",
      "\n",
      "Iteration: 76000, Loss: 15.377773\n",
      "\n",
      "Pnarmant\n",
      "Yupaum\n",
      "Pun\n",
      "Pulrant\n",
      "Veegsh\n",
      "Inja\n",
      "Eena\n",
      "Oswarinat\n",
      "Mulasha\n",
      "Ashyamyee\n",
      "\n",
      "\n",
      "Iteration: 78000, Loss: 15.260911\n",
      "\n",
      "Ubaldu\n",
      "Baper\n",
      "Auzabana\n",
      "Robin\n",
      "Poshomi\n",
      "Ayamo\n",
      "Aulisha\n",
      "Jhayash\n",
      "Oshash\n",
      "Eend\n",
      "\n",
      "\n",
      "Iteration: 80000, Loss: 15.530162\n",
      "\n",
      "Jeesm\n",
      "Seshuta\n",
      "Bhanjaa\n",
      "Bhin\n",
      "Sangan\n",
      "Awarikssa\n",
      "Ustit\n",
      "Bipam\n",
      "Utya\n",
      "Agmish\n",
      "\n",
      "\n",
      "Iteration: 82000, Loss: 15.556394\n",
      "\n",
      "Ogee\n",
      "Uschhal\n",
      "Rozndresh\n",
      "Poobindam\n",
      "Yittan\n",
      "Akchan\n",
      "Yasmin\n",
      "Axen\n",
      "Raksook\n",
      "Bishwor\n",
      "\n",
      "\n",
      "Iteration: 84000, Loss: 15.347838\n",
      "\n",
      "Prasoj\n",
      "Juna\n",
      "Sudshar\n",
      "Aisuna\n",
      "Rem\n",
      "Enj\n",
      "Divnop\n",
      "Gural\n",
      "Eeer\n",
      "Jagasha\n",
      "\n",
      "\n",
      "Iteration: 86000, Loss: 15.275874\n",
      "\n",
      "Kee\n",
      "Shashrija\n",
      "Shevan\n",
      "Ayishati\n",
      "Vambha\n",
      "Ayish\n",
      "Shis\n",
      "Baman\n",
      "Shilikajusa\n",
      "Vebha\n",
      "\n",
      "\n",
      "Iteration: 88000, Loss: 15.503609\n",
      "\n",
      "Suvita\n",
      "Ubin\n",
      "Ayu\n",
      "Vinana\n",
      "Shivu\n",
      "Aulendra\n",
      "Shobesh\n",
      "Bindos\n",
      "Budhundree\n",
      "Rilon\n",
      "\n",
      "\n",
      "Iteration: 90000, Loss: 15.525616\n",
      "\n",
      "Abich\n",
      "Ejitan\n",
      "Sisama\n",
      "Biyanik\n",
      "Bitib\n",
      "Enina\n",
      "Urpali\n",
      "Meskon\n",
      "Ubalani\n",
      "Shil\n",
      "\n",
      "\n",
      "Iteration: 92000, Loss: 15.311973\n",
      "\n",
      "Rajman\n",
      "Achishi\n",
      "Onik\n",
      "Bishum\n",
      "Banidhan\n",
      "Juprissinikra\n",
      "Ilurak\n",
      "Ikshanu\n",
      "Bikisani\n",
      "Utram\n",
      "\n",
      "\n",
      "Iteration: 94000, Loss: 15.250997\n",
      "\n",
      "Urat\n",
      "Ailechi\n",
      "Anasna\n",
      "Ujweshwa\n",
      "Dhazna\n",
      "Ushuma\n",
      "Rugu\n",
      "Phusam\n",
      "Omari\n",
      "Ubhark\n",
      "\n",
      "\n",
      "Iteration: 96000, Loss: 15.517277\n",
      "\n",
      "Ikeshwor\n",
      "Adur\n",
      "Dundratha\n",
      "Ruse\n",
      "Praneba\n",
      "Uvit\n",
      "Byayesh\n",
      "Uchreela\n",
      "Epala\n",
      "Knatangya\n",
      "\n",
      "\n",
      "Iteration: 98000, Loss: 15.498104\n",
      "\n",
      "Rusirabithan\n",
      "Biphat\n",
      "Purminita\n",
      "Bithiran\n",
      "Eshwit\n",
      "Jenish\n",
      "Vasesh\n",
      "Bhuden\n",
      "Bichendra\n",
      "Bigsaten\n",
      "\n",
      "\n",
      "Iteration: 100000, Loss: 15.322775\n",
      "\n",
      "Sotum\n",
      "Anju\n",
      "Sumab\n",
      "Rashati\n",
      "Rekak\n",
      "Sjarig\n",
      "Ganika\n",
      "Ammanae\n",
      "Tretar\n",
      "Sovesh\n",
      "\n",
      "\n",
      "Iteration: 102000, Loss: 15.241426\n",
      "\n",
      "Gomit\n",
      "Bhashin\n",
      "Ajan\n",
      "Aupu\n",
      "Aamhop\n",
      "Uha\n",
      "Jushu\n",
      "Isham\n",
      "Subinramaash\n",
      "Leshan\n",
      "\n",
      "\n",
      "Iteration: 104000, Loss: 15.466123\n",
      "\n",
      "Vindrakang\n",
      "Rebekshes\n",
      "Pushni\n",
      "Divishga\n",
      "Devik\n",
      "Binad\n",
      "Sabh\n",
      "Mola\n",
      "Bitika\n",
      "Eswar\n",
      "\n",
      "\n",
      "Iteration: 106000, Loss: 15.444018\n",
      "\n",
      "Subish\n",
      "Eshundu\n",
      "Jhanitan\n",
      "Ureki\n",
      "Sugyare\n",
      "Dhuvetan\n",
      "Rajam\n",
      "Bishra\n",
      "Bipish\n",
      "Suchan\n",
      "\n",
      "\n",
      "Iteration: 108000, Loss: 15.292746\n",
      "\n",
      "Agrukashan\n",
      "Rikeshwar\n",
      "Pranitaraj\n",
      "Yeekashworla\n",
      "Urti\n",
      "Yunan\n",
      "Igan\n",
      "Iksha\n",
      "Gyabliya\n",
      "Uvincowari\n",
      "\n",
      "\n",
      "Iteration: 110000, Loss: 15.285444\n",
      "\n",
      "Pun\n",
      "Jitharshharbama\n",
      "Ishti\n",
      "Kegit\n",
      "Samreyogi\n",
      "Aipan\n",
      "Gubesh\n",
      "Bikarshya\n",
      "Onchath\n",
      "Drishut\n",
      "\n",
      "\n",
      "Iteration: 112000, Loss: 15.425474\n",
      "\n",
      "Bina\n",
      "Balchrishbi\n",
      "Eeraavir\n",
      "Makartu\n",
      "Rantib\n",
      "Begwa\n",
      "Engti\n",
      "Uremna\n",
      "Uprsur\n",
      "Rimpalbi\n",
      "\n",
      "\n",
      "Iteration: 114000, Loss: 15.439120\n",
      "\n",
      "Utunam\n",
      "Eelen\n",
      "Arikhawal\n",
      "Ireshwah\n",
      "Asundra\n",
      "Eeesh\n",
      "Rhusta\n",
      "Javarkal\n",
      "Binjant\n",
      "Eecsha\n",
      "\n",
      "\n",
      "Iteration: 116000, Loss: 15.348496\n",
      "\n",
      "Sumasha\n",
      "Aleshma\n",
      "Sunushri\n",
      "Ulju\n",
      "Chhrij\n",
      "Prajaj\n",
      "Beshma\n",
      "Anujensan\n",
      "Unga\n",
      "Pranash\n",
      "\n",
      "\n",
      "Iteration: 118000, Loss: 15.295688\n",
      "\n",
      "Bannadha\n",
      "Umsha\n",
      "Amon\n",
      "Oxnsha\n",
      "Chshaganyashan\n",
      "Bharshan\n",
      "Udao\n",
      "Sumroma\n",
      "Urvita\n",
      "Susiri\n",
      "\n",
      "\n",
      "Iteration: 120000, Loss: 15.413973\n",
      "\n",
      "Aurath\n",
      "Inju\n",
      "Shivak\n",
      "Rubnki\n",
      "Ithamat\n",
      "Rojal\n",
      "Gsheswa\n",
      "Ugoni\n",
      "Suiva\n",
      "Shina\n",
      "\n",
      "\n",
      "Iteration: 122000, Loss: 15.430792\n",
      "\n",
      "Rizad\n",
      "Rayan\n",
      "Shatit\n",
      "Bitnu\n",
      "Ejan\n",
      "Ohedhik\n",
      "Marvina\n",
      "Eesujay\n",
      "Paw\n",
      "Dagbe\n",
      "\n",
      "\n",
      "Iteration: 124000, Loss: 15.321201\n",
      "\n",
      "Ugya\n",
      "Pumal\n",
      "Binteb\n",
      "Bah\n",
      "Jubala\n",
      "Huyelva\n",
      "Uashal\n",
      "Eenu\n",
      "Mudim\n",
      "Alashanji\n",
      "\n",
      "\n",
      "Iteration: 126000, Loss: 15.281380\n",
      "\n",
      "Ruti\n",
      "Anurshira\n",
      "Matri\n",
      "Undha\n",
      "Ugona\n",
      "Bisushi\n",
      "Ushakarandra\n",
      "Unchishya\n",
      "Rajjan\n",
      "Nimur\n",
      "\n",
      "\n",
      "Iteration: 128000, Loss: 15.410249\n",
      "\n",
      "Aubalek\n",
      "Ubekesh\n",
      "Urendr\n",
      "Bhabyanda\n",
      "Rogla\n",
      "Bigked\n",
      "Geosh\n",
      "Jeshemi\n",
      "Utmath\n",
      "Bili\n",
      "\n",
      "\n",
      "Iteration: 130000, Loss: 15.367537\n",
      "\n",
      "Bhawendra\n",
      "Eliva\n",
      "Ayulam\n",
      "Bippayav\n",
      "Vevlshagjiti\n",
      "Veerind\n",
      "Eerdel\n",
      "Ahhim\n",
      "Sutith\n",
      "Renil\n",
      "\n",
      "\n",
      "Iteration: 132000, Loss: 15.283044\n",
      "\n",
      "Aushinm\n",
      "Seshawa\n",
      "Huryon\n",
      "Reecni\n",
      "Jeeb\n",
      "Jisthanu\n",
      "Bhrijdep\n",
      "Bhubrandrika\n",
      "Swikshothas\n",
      "Minav\n",
      "\n",
      "\n",
      "Iteration: 134000, Loss: 15.251558\n",
      "\n",
      "Bushumar\n",
      "Jeana\n",
      "Rusnit\n",
      "U\n",
      "Ushwar\n",
      "Uroz\n",
      "Ramandea\n",
      "Bulusha\n",
      "Iman\n",
      "Jishor\n",
      "\n",
      "\n",
      "Iteration: 136000, Loss: 15.369700\n",
      "\n",
      "Asat\n",
      "Daiteo\n",
      "Bishma\n",
      "Etna\n",
      "Alesh\n",
      "Ukral\n",
      "Tani\n",
      "Sujou\n",
      "Shaminan\n",
      "Beela\n",
      "\n",
      "\n",
      "Iteration: 138000, Loss: 15.353657\n",
      "\n",
      "Aghitr\n",
      "Zeeraja\n",
      "Chhit\n",
      "Pesh\n",
      "Siyang\n",
      "Eemu\n",
      "Untshal\n",
      "Nilesh\n",
      "Bipas\n",
      "Regang\n",
      "\n",
      "\n",
      "Iteration: 140000, Loss: 15.316629\n",
      "\n",
      "Onsinay\n",
      "Pabbiv\n",
      "Schita\n",
      "Ishida\n",
      "Agileya\n",
      "Anyanwan\n",
      "Heehked\n",
      "Jupita\n",
      "Ragiuk\n",
      "Ongas\n",
      "\n",
      "\n",
      "Iteration: 142000, Loss: 15.236399\n",
      "\n",
      "Rupadejan\n",
      "Ripasani\n",
      "Bararanjee\n",
      "Ukja\n",
      "Sand\n",
      "Binas\n",
      "Shanisa\n",
      "Rencha\n",
      "Sapin\n",
      "Eeta\n",
      "\n",
      "\n",
      "Iteration: 144000, Loss: 15.328076\n",
      "\n",
      "Ateshmana\n",
      "Rumsan\n",
      "Kumray\n",
      "Bisham\n",
      "Bhabinan\n",
      "Shrika\n",
      "Ujyat\n",
      "Manilta\n",
      "Uamrsheswara\n",
      "Unjad\n",
      "\n",
      "\n",
      "Iteration: 146000, Loss: 15.385829\n",
      "\n",
      "Endrab\n",
      "Rchbi\n",
      "Rushekhna\n",
      "Sujijenish\n",
      "Silash\n",
      "Anin\n",
      "Viway\n",
      "Romina\n",
      "Bindesh\n",
      "Sumi\n",
      "\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration: 148000, Loss: 15.290607\n",
      "\n",
      "Ranika\n",
      "Aarajang\n",
      "Usura\n",
      "Jewano\n",
      "Aasiy\n",
      "Aashchar\n",
      "Pangitada\n",
      "Junshanw\n",
      "Umyal\n",
      "Soes\n",
      "\n",
      "\n",
      "Iteration: 150000, Loss: 15.244997\n",
      "\n",
      "Vartanchda\n",
      "Anarsha\n",
      "Almalal\n",
      "Anshehaban\n",
      "Viranga\n",
      "Bprizel\n",
      "Shimana\n",
      "Araj\n",
      "Pratendrajasam\n",
      "Jeali\n",
      "\n",
      "\n",
      "Iteration: 152000, Loss: 15.332107\n",
      "\n",
      "Anrish\n",
      "Enaswer\n",
      "An\n",
      "Runal\n",
      "Rvajina\n",
      "Eetalt\n",
      "Bikina\n",
      "Aili\n",
      "Myat\n",
      "Rani\n",
      "\n",
      "\n",
      "Iteration: 154000, Loss: 15.327888\n",
      "\n",
      "Brijesh\n",
      "Rima\n",
      "Darukoya\n",
      "Man\n",
      "Biswajudap\n",
      "Bipeasheh\n",
      "Aasheleek\n",
      "Sokayan\n",
      "Ukhdeeshlaa\n",
      "Mudash\n",
      "\n",
      "\n",
      "Iteration: 156000, Loss: 15.297226\n",
      "\n",
      "Jupesh\n",
      "Sutparh\n",
      "Mahis\n",
      "Iridis\n",
      "Ranjew\n",
      "Junda\n",
      "Raghishar\n",
      "Aasmalth\n",
      "Meni\n",
      "Keshwohhal\n",
      "\n",
      "\n",
      "Iteration: 158000, Loss: 15.244905\n",
      "\n",
      "Rnatii\n",
      "Urmesya\n",
      "Enesba\n",
      "Abhidyu\n",
      "Rujansa\n",
      "Jilea\n",
      "Ushriban\n",
      "Cefwal\n",
      "Cadarman\n",
      "Ashan\n",
      "\n",
      "\n",
      "Iteration: 160000, Loss: 15.322017\n",
      "\n",
      "Dishit\n",
      "Bikritti\n",
      "Gurjal\n",
      "Biju\n",
      "Uramanti\n",
      "Awhitendit\n",
      "Achan\n",
      "Aapantaav\n",
      "Srgeen\n",
      "Rishit\n",
      "\n",
      "\n",
      "Iteration: 162000, Loss: 15.324701\n",
      "\n",
      "Roolesh\n",
      "Ahadap\n",
      "Sumana\n",
      "Aural\n",
      "Slisu\n",
      "Heepati\n",
      "Uhi\n",
      "Jukeya\n",
      "Aabhad\n",
      "Smanash\n",
      "\n",
      "\n",
      "Iteration: 164000, Loss: 15.271274\n",
      "\n",
      "Huragee\n",
      "Ghala\n",
      "Virina\n",
      "Eulad\n",
      "Auramayakinda\n",
      "Aroj\n",
      "Vishyarshuk\n",
      "Sanshal\n",
      "Minikriyaji\n",
      "Ranjira\n",
      "\n",
      "\n",
      "Iteration: 166000, Loss: 15.225551\n",
      "\n",
      "Lusudantrit\n",
      "Biugab\n",
      "Britu\n",
      "Kndhisharans\n",
      "Ukhim\n",
      "Sojba\n",
      "Rujxat\n",
      "Urghas\n",
      "Imya\n",
      "Bhamana\n",
      "\n",
      "\n",
      "Iteration: 168000, Loss: 15.283562\n",
      "\n",
      "Vaharisan\n",
      "Rie\n",
      "Charab\n",
      "Yukansatina\n",
      "Surit\n",
      "Biragai\n",
      "Birait\n",
      "Prina\n",
      "Dibam\n",
      "Urmana\n",
      "\n",
      "\n",
      "Iteration: 170000, Loss: 15.332833\n",
      "\n",
      "Nivend\n",
      "Ushoum\n",
      "Jedikshh\n",
      "Anufar\n",
      "Nismith\n",
      "Melkaritha\n",
      "Uepin\n",
      "Utgari\n",
      "Bidhelita\n",
      "Sadheenc\n",
      "\n",
      "\n",
      "Iteration: 172000, Loss: 15.292369\n",
      "\n",
      "Biroshana\n",
      "Aepni\n",
      "Aibika\n",
      "Umrima\n",
      "Sagin\n",
      "Biazyoniura\n",
      "Vishirda\n",
      "Ramikla\n",
      "Mangisha\n",
      "Baka\n",
      "\n",
      "\n",
      "Iteration: 174000, Loss: 15.216265\n",
      "\n",
      "Uransha\n",
      "Arin\n",
      "Ugad\n",
      "Yugar\n",
      "Umichna\n",
      "Nipru\n",
      "Rajisan\n",
      "Hardesman\n",
      "Cyekchan\n",
      "Bhupama\n",
      "\n",
      "\n",
      "Iteration: 176000, Loss: 15.271903\n",
      "\n",
      "Mir\n",
      "Beshya\n",
      "Preeya\n",
      "Ruptish\n",
      "Rastsh\n",
      "Ratjani\n",
      "Rashri\n",
      "Een\n",
      "Anaraoj\n",
      "Aisaksha\n",
      "\n",
      "\n",
      "Iteration: 178000, Loss: 15.313420\n",
      "\n",
      "Ayeek\n",
      "Jeeran\n",
      "Epishava\n",
      "Udhen\n",
      "Uudhekh\n",
      "Sheshna\n",
      "Yuchet\n",
      "Kheukawankaryaram\n",
      "Jangich\n",
      "Orakra\n",
      "\n",
      "\n",
      "Iteration: 180000, Loss: 15.300515\n",
      "\n",
      "Rubhush\n",
      "Urit\n",
      "Bergunshagh\n",
      "Ohatu\n",
      "Jeva\n",
      "Udyal\n",
      "Jiti\n",
      "Obham\n",
      "Ampu\n",
      "Byad\n",
      "\n",
      "\n",
      "Iteration: 182000, Loss: 15.159960\n",
      "\n",
      "Beekh\n",
      "Kusoma\n",
      "Masescharari\n",
      "Mesh\n",
      "Bud\n",
      "Raswison\n",
      "Mukakaniteet\n",
      "Nibul\n",
      "Maaras\n",
      "Usilaram\n",
      "\n",
      "\n",
      "Iteration: 184000, Loss: 15.296460\n",
      "\n",
      "Bhitika\n",
      "Biswas\n",
      "Rikanta\n",
      "Suvek\n",
      "Joona\n",
      "Bharunani\n",
      "Birtika\n",
      "Susankal\n",
      "Seswa\n",
      "Smeep\n",
      "\n",
      "\n",
      "Iteration: 186000, Loss: 15.312625\n",
      "\n",
      "Suchil\n",
      "Rehpama\n",
      "Yaghori\n",
      "Fajil\n",
      "Dheshor\n",
      "Mankaderi\n",
      "Sudyab\n",
      "Ashindra\n",
      "Jeep\n",
      "Ukhmesh\n",
      "\n",
      "\n",
      "Iteration: 188000, Loss: 15.268505\n",
      "\n",
      "Samba\n",
      "Pambdrhyith\n",
      "Ratij\n",
      "Urkub\n",
      "Bapraj\n",
      "Hesanta\n",
      "Hamanjamani\n",
      "Jonesh\n",
      "Bimbaki\n",
      "Pubbaz\n",
      "\n",
      "\n",
      "Iteration: 190000, Loss: 15.195630\n",
      "\n",
      "Aunudha\n",
      "Ayodhu\n",
      "Ahanku\n",
      "Aalpag\n",
      "Achhuma\n",
      "Uzen\n",
      "Aushika\n",
      "Uhenal\n",
      "Bipaswar\n",
      "Ashayu\n",
      "\n",
      "\n",
      "Iteration: 192000, Loss: 15.318417\n",
      "\n",
      "Bishurti\n",
      "Babit\n",
      "Eewina\n",
      "Pulbar\n",
      "Migeena\n",
      "Rumini\n",
      "Jaidradhan\n",
      "Akanalojha\n",
      "Bipashrihan\n",
      "Oovipsha\n",
      "\n",
      "\n",
      "Iteration: 194000, Loss: 15.306352\n",
      "\n",
      "Aovibi\n",
      "Oapin\n",
      "Peshthur\n",
      "Jages\n",
      "Urjan\n",
      "Ahish\n",
      "Jupisan\n",
      "Nabish\n",
      "Ayantol\n",
      "Aakar\n",
      "\n",
      "\n",
      "Iteration: 196000, Loss: 15.217971\n",
      "\n",
      "Vanar\n",
      "Banush\n",
      "Rasana\n",
      "Arubu\n",
      "Rupan\n",
      "Iben\n",
      "Matit\n",
      "Ganansha\n",
      "Ashan\n",
      "Anoj\n",
      "\n",
      "\n",
      "Iteration: 198000, Loss: 15.205749\n",
      "\n",
      "Eec\n",
      "Aalunsha\n",
      "Mukenca\n",
      "Vanup\n",
      "Ohanendra\n",
      "Khalma\n",
      "Baniz\n",
      "Uriti\n",
      "Urajtaijan\n",
      "Ritel\n",
      "\n",
      "\n",
      "Iteration: 200000, Loss: 15.331148\n",
      "\n",
      "Sarishya\n",
      "Satoos\n",
      "Manin\n",
      "Jabtya\n",
      "Bhutnak\n",
      "Baroj\n",
      "Bhuwaranat\n",
      "Unjaya\n",
      "Swarmmanda\n",
      "Rachhshalaman\n",
      "\n",
      "\n",
      "Iteration: 202000, Loss: 15.274433\n",
      "\n",
      "Paksha\n",
      "Uhma\n",
      "Sanjila\n",
      "Urajant\n",
      "Uadswal\n",
      "Eeghna\n",
      "Julin\n",
      "Ooshadhari\n",
      "Aanit\n",
      "Arasunda\n",
      "\n",
      "\n",
      "Iteration: 204000, Loss: 15.214065\n",
      "\n",
      "Deveb\n",
      "Ramsingan\n",
      "Jama\n",
      "Arah\n",
      "Sirbin\n",
      "Syok\n",
      "Mordip\n",
      "Ropandhana\n",
      "Birbini\n",
      "Buroya\n",
      "\n",
      "\n",
      "Iteration: 206000, Loss: 15.187881\n",
      "\n",
      "Shajenddra\n",
      "Bishma\n",
      "Rayaj\n",
      "Muban\n",
      "Ejena\n",
      "Bitush\n",
      "Uman\n",
      "Sasha\n",
      "Abitrai\n",
      "Juban\n",
      "\n",
      "\n",
      "Iteration: 208000, Loss: 15.410824\n",
      "\n",
      "Unjay\n",
      "Ahitami\n",
      "Eshar\n",
      "Rusun\n",
      "Rajom\n",
      "Eela\n",
      "Eenab\n",
      "Mavis\n",
      "Remanand\n",
      "Sabindra\n",
      "\n",
      "\n",
      "Iteration: 210000, Loss: 15.277947\n",
      "\n",
      "Susina\n",
      "Sukesh\n",
      "Ukina\n",
      "Ahopan\n",
      "Biddya\n",
      "Amush\n",
      "Aushstha\n",
      "Arbhis\n",
      "Usmi\n",
      "Aruyas\n",
      "\n",
      "\n",
      "Iteration: 212000, Loss: 15.185487\n",
      "\n",
      "Ujaj\n",
      "Ushid\n",
      "Bhashu\n",
      "Pikhaman\n",
      "Pursa\n",
      "Usik\n",
      "Jupra\n",
      "Jewendr\n",
      "Chinshribya\n",
      "Rienda\n",
      "\n",
      "\n",
      "Iteration: 214000, Loss: 15.235656\n",
      "\n",
      "Saiba\n",
      "Galrat\n",
      "Ushkai\n",
      "Yutaun\n",
      "Udhamal\n",
      "Bayanin\n",
      "Rasam\n",
      "Jutenh\n",
      "Gubula\n",
      "Uhani\n",
      "\n",
      "\n",
      "Iteration: 216000, Loss: 15.399272\n",
      "\n",
      "Charsadha\n",
      "Divina\n",
      "Yamroher\n",
      "Aereena\n",
      "Ehisan\n",
      "Adikarim\n",
      "Umeahlajna\n",
      "Achchyudha\n",
      "Urinev\n",
      "Byushkuth\n",
      "\n",
      "\n",
      "Iteration: 218000, Loss: 15.252730\n",
      "\n",
      "Bidhik\n",
      "Gyascha\n",
      "Usvina\n",
      "Chhitha\n",
      "Uelak\n",
      "Mina\n",
      "Anupam\n",
      "Uelan\n",
      "Biena\n",
      "Buthan\n",
      "\n",
      "\n",
      "Iteration: 220000, Loss: 15.160161\n",
      "\n",
      "Rumana\n",
      "Divanali\n",
      "Uron\n",
      "Barita\n",
      "Uites\n",
      "Iniya\n",
      "Romiti\n",
      "Eeta\n",
      "Urita\n",
      "Binee\n",
      "\n",
      "\n",
      "Iteration: 222000, Loss: 15.157176\n",
      "\n",
      "Juhmap\n",
      "Rachan\n",
      "Cutar\n",
      "Eny\n",
      "Ausha\n",
      "Bimos\n",
      "Radeem\n",
      "Urota\n",
      "Maheshwar\n",
      "Rohjijwar\n",
      "\n",
      "\n",
      "Iteration: 224000, Loss: 15.350050\n",
      "\n",
      "Divibja\n",
      "Mish\n",
      "Kponushadh\n",
      "Birita\n",
      "Aarona\n",
      "Krithjud\n",
      "Bekrav\n",
      "Sunau\n",
      "Uyahma\n",
      "Ushibherama\n",
      "\n",
      "\n",
      "Iteration: 226000, Loss: 15.227473\n",
      "\n",
      "Mooshmantha\n",
      "Suhant\n",
      "Bibharpan\n",
      "Ishw\n",
      "Chand\n",
      "Urak\n",
      "Murah\n",
      "Obirina\n",
      "Irla\n",
      "Ujundra\n",
      "\n",
      "\n",
      "Iteration: 228000, Loss: 15.190091\n",
      "\n",
      "Uukesh\n",
      "Uryahma\n",
      "Rachh\n",
      "Putraanair\n",
      "Sripanti\n",
      "Ravikika\n",
      "Uren\n",
      "Asarina\n",
      "Minam\n",
      "Shirku\n",
      "\n",
      "\n",
      "Iteration: 230000, Loss: 15.137096\n",
      "\n",
      "Itansha\n",
      "Ritan\n",
      "Mokta\n",
      "Kpila\n",
      "Rahanra\n",
      "Hiraj\n",
      "Suvan\n",
      "Uev\n",
      "Anjona\n",
      "Royanji\n",
      "\n",
      "\n",
      "Iteration: 232000, Loss: 15.336053\n",
      "\n",
      "Ohan\n",
      "Umesh\n",
      "Valisha\n",
      "Opira\n",
      "Hetu\n",
      "Destilaz\n",
      "Udhartita\n",
      "Rizan\n",
      "Uram\n",
      "Urnita\n",
      "\n",
      "\n",
      "Iteration: 234000, Loss: 15.196730\n",
      "\n",
      "Chit\n",
      "Gogmat\n",
      "Rabhan\n",
      "Oovar\n",
      "Aurtsa\n",
      "Padik\n",
      "Krishak\n",
      "Uka\n",
      "Aripsa\n",
      "Sumkit\n",
      "\n",
      "\n",
      "Iteration: 236000, Loss: 15.171882\n",
      "\n",
      "Briwarag\n",
      "Uabhu\n",
      "Eizena\n",
      "Rajin\n",
      "Auman\n",
      "Prasha\n",
      "Anam\n",
      "Eshuthab\n",
      "Revishi\n",
      "Sa\n",
      "\n",
      "\n",
      "Iteration: 238000, Loss: 15.184160\n",
      "\n",
      "Ineshwari\n",
      "Hegti\n",
      "Umar\n",
      "Sarupanjayalka\n",
      "Malsuchhu\n",
      "Uusz\n",
      "Maharush\n",
      "Jhala\n",
      "Sarwarint\n",
      "Runu\n",
      "\n",
      "\n",
      "Iteration: 240000, Loss: 15.322983\n",
      "\n",
      "Bhagan\n",
      "Aunend\n",
      "Sarosbi\n",
      "Arjana\n",
      "Mita\n",
      "Binina\n",
      "Ghamot\n",
      "Uteva\n",
      "Ita\n",
      "Sukandray\n",
      "\n",
      "\n",
      "Iteration: 242000, Loss: 15.136922\n",
      "\n",
      "Ahbipal\n",
      "Rmandiv\n",
      "Manta\n",
      "Shitkar\n",
      "Ufan\n",
      "Ayunjan\n",
      "Raksha\n",
      "Ojat\n",
      "Bindhh\n",
      "Jatil\n",
      "\n",
      "\n",
      "Iteration: 244000, Loss: 15.178117\n",
      "\n",
      "Arigrabha\n",
      "Dayosu\n",
      "Swooc\n",
      "Urisha\n",
      "Bhavansh\n",
      "Ishan\n",
      "Virikatito\n",
      "Jagedhrik\n",
      "Rupaphant\n",
      "Urikshan\n",
      "\n",
      "\n",
      "Iteration: 246000, Loss: 15.109911\n",
      "\n",
      "Ainsam\n",
      "Ramabhikanana\n",
      "Kulisha\n",
      "Riseska\n",
      "Eohsnira\n",
      "Mushil\n",
      "Aailev\n",
      "Eem\n",
      "Antehyal\n",
      "Ishirali\n",
      "\n",
      "\n",
      "Iteration: 248000, Loss: 15.296557\n",
      "\n",
      "Bbinjeyi\n",
      "Uhina\n",
      "Birini\n",
      "Prepesth\n",
      "Chirubi\n",
      "Aajreni\n",
      "Shigon\n",
      "Puna\n",
      "Baureet\n",
      "Bilep\n",
      "\n",
      "\n",
      "Iteration: 250000, Loss: 15.155540\n",
      "\n",
      "Aasendri\n",
      "Mahaaba\n",
      "Uelk\n",
      "Inira\n",
      "Manundi\n",
      "Naneerath\n",
      "Ranuka\n",
      "Ramhand\n",
      "Arajad\n",
      "Ageep\n",
      "\n",
      "\n",
      "Iteration: 252000, Loss: 15.185223\n",
      "\n",
      "Adomamna\n",
      "Rujen\n",
      "Shtrava\n",
      "Iriya\n",
      "Didesh\n",
      "Ashewat\n",
      "Uhajichha\n",
      "Nanung\n",
      "Ahena\n",
      "Riti\n",
      "\n",
      "\n",
      "Iteration: 254000, Loss: 15.087961\n",
      "\n",
      "Rishisha\n",
      "Uhot\n",
      "Ish\n",
      "Jiltodhart\n",
      "Sungornahar\n",
      "Jeet\n",
      "Jiy\n",
      "Uhar\n",
      "Pri\n",
      "Enil\n",
      "\n",
      "\n",
      "Iteration: 256000, Loss: 15.292601\n",
      "\n",
      "Jigjibh\n",
      "Raveka\n",
      "Banoj\n",
      "Diveta\n",
      "Kshikratriny\n",
      "Aalish\n",
      "Aarojan\n",
      "Uddireku\n",
      "Usush\n",
      "Uheki\n",
      "\n",
      "\n",
      "Iteration: 258000, Loss: 15.114043\n",
      "\n",
      "Ganet\n",
      "Uaypir\n",
      "Urispla\n",
      "Ranunal\n",
      "Aarana\n",
      "Ivesw\n",
      "Rakman\n",
      "Usyav\n",
      "Manij\n",
      "Ayantorsaakihar\n",
      "\n",
      "\n",
      "Iteration: 260000, Loss: 15.155157\n",
      "\n",
      "I\n",
      "Jeyanish\n",
      "Urishma\n",
      "Ibinshak\n",
      "Dishrita\n",
      "Urish\n",
      "Bayanis\n",
      "Jaoju\n",
      "Ragita\n",
      "Ujoti\n",
      "\n",
      "\n",
      "Iteration: 262000, Loss: 15.094522\n",
      "\n",
      "Bhiv\n",
      "Igay\n",
      "Asechh\n",
      "Rikal\n",
      "Bipan\n",
      "Ohrat\n",
      "Ruppyashri\n",
      "Usadha\n",
      "Rupasip\n",
      "Ranita\n",
      "\n",
      "\n",
      "Iteration: 264000, Loss: 15.269480\n",
      "\n",
      "Fanjika\n",
      "Amyaash\n",
      "Uujupal\n",
      "Eebendra\n",
      "Rahal\n",
      "Urim\n",
      "Rum\n",
      "Urshan\n",
      "Bantik\n",
      "Crishna\n",
      "\n",
      "\n",
      "Iteration: 266000, Loss: 15.090869\n",
      "\n",
      "Ugantha\n",
      "Sumeshemsa\n",
      "Srkrami\n",
      "Mangah\n",
      "Nmitam\n",
      "Mahlena\n",
      "Miheshwa\n",
      "Brisman\n",
      "Bavsutaragan\n",
      "Prishmaha\n",
      "\n",
      "\n",
      "Iteration: 268000, Loss: 15.187907\n",
      "\n",
      "Ikratitha\n",
      "Bimiti\n",
      "Nimimba\n",
      "Uraswal\n",
      "Saimal\n",
      "Uraswon\n",
      "Ahikes\n",
      "Urmija\n",
      "Biricesha\n",
      "Imlan\n",
      "\n",
      "\n",
      "Iteration: 270000, Loss: 15.130111\n",
      "\n",
      "Andushanaala\n",
      "Aeroj\n",
      "Vanfumjan\n",
      "Radad\n",
      "Aijanaau\n",
      "Ramu\n",
      "Sathanu\n",
      "Rupasanan\n",
      "Basekh\n",
      "Munat\n",
      "\n",
      "\n",
      "Iteration: 272000, Loss: 15.273368\n",
      "\n",
      "Udawalir\n",
      "Binasu\n",
      "Ailikyum\n",
      "Bidush\n",
      "Aaushwanam\n",
      "Bishbu\n",
      "Niyem\n",
      "Jauzat\n",
      "Alem\n",
      "Ravendra\n",
      "\n",
      "\n",
      "Iteration: 274000, Loss: 15.103595\n",
      "\n",
      "Bibhan\n",
      "Rini\n",
      "Somen\n",
      "Diyaks\n",
      "Ouhalan\n",
      "Gakala\n",
      "Jhishkri\n",
      "Amri\n",
      "Manish\n",
      "Uub\n",
      "\n",
      "\n",
      "Iteration: 276000, Loss: 15.235330\n",
      "\n",
      "Ramaj\n",
      "Arbha\n",
      "Bipesh\n",
      "Afhisha\n",
      "Yasil\n",
      "Subranka\n",
      "Urapprat\n",
      "Punw\n",
      "Uebin\n",
      "Imnya\n",
      "\n",
      "\n",
      "Iteration: 278000, Loss: 15.179062\n",
      "\n",
      "Urillesh\n",
      "Eelsok\n",
      "Amam\n",
      "Auchash\n",
      "Uharth\n",
      "Dima\n",
      "Abhesab\n",
      "Uraju\n",
      "Aanchan\n",
      "Roneena\n",
      "\n",
      "\n",
      "Iteration: 280000, Loss: 15.255522\n",
      "\n",
      "Rjeamaj\n",
      "Rakrita\n",
      "Uryoj\n",
      "Ram\n",
      "Charbal\n",
      "Umbal\n",
      "Juwantand\n",
      "Suwar\n",
      "Reni\n",
      "Rajhal\n",
      "\n",
      "\n",
      "Iteration: 282000, Loss: 15.114400\n",
      "\n",
      "Bikshyy\n",
      "Sushmail\n",
      "Uooman\n",
      "Aabikar\n",
      "Pbendraya\n",
      "Aayan\n",
      "Bhadil\n",
      "Ayusuki\n",
      "Sin\n",
      "Tovshyu\n",
      "\n",
      "\n",
      "Iteration: 284000, Loss: 15.207572\n",
      "\n",
      "Uupar\n",
      "Mogees\n",
      "Yuteera\n",
      "Mayajarindra\n",
      "Samshbha\n",
      "Uuden\n",
      "Ika\n",
      "Reshab\n",
      "Yupan\n",
      "Ugila\n",
      "\n",
      "\n",
      "Iteration: 286000, Loss: 15.125519\n",
      "\n",
      "Priuja\n",
      "Bicitri\n",
      "Fiema\n",
      "Gena\n",
      "Uhomar\n",
      "Auraj\n",
      "Eepasa\n",
      "Ripash\n",
      "Manji\n",
      "Jekkar\n",
      "\n",
      "\n",
      "Iteration: 288000, Loss: 15.243126\n",
      "\n",
      "Shujanara\n",
      "Praukabin\n",
      "Joras\n",
      "Urila\n",
      "Rana\n",
      "Rashish\n",
      "Benam\n",
      "Shurampra\n",
      "Skronandrav\n",
      "Uamun\n",
      "\n",
      "\n",
      "Iteration: 290000, Loss: 15.071158\n",
      "\n",
      "Aanpdhan\n",
      "Awat\n",
      "Uncrshan\n",
      "Vipsan\n",
      "Aayan\n",
      "Aanidhan\n",
      "Sokhor\n",
      "Shawunt\n",
      "Izen\n",
      "Eerob\n",
      "\n",
      "\n",
      "Iteration: 292000, Loss: 15.243623\n",
      "\n",
      "Iglendra\n",
      "Ufeel\n",
      "Ramej\n",
      "Kriyakee\n",
      "Dheefun\n",
      "Amotra\n",
      "Deshovasikana\n",
      "Uhina\n",
      "Prashaz\n",
      "Shijan\n",
      "\n",
      "\n",
      "Iteration: 294000, Loss: 15.102875\n",
      "\n",
      "Udchanjan\n",
      "Bhuda\n",
      "Sanjeeti\n",
      "Gagashra\n",
      "Uhadev\n",
      "Eras\n",
      "Peshryula\n",
      "Ayraj\n",
      "Aamrnandep\n",
      "Ithan\n",
      "\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration: 296000, Loss: 15.223631\n",
      "\n",
      "Krilakh\n",
      "Mishrita\n",
      "Uhiv\n",
      "Etanandhara\n",
      "Birin\n",
      "Uban\n",
      "Gunkaha\n",
      "Ikdan\n",
      "Ruchanta\n",
      "Aalir\n",
      "\n",
      "\n",
      "Iteration: 298000, Loss: 15.095909\n",
      "\n",
      "Ouk\n",
      "Ujan\n",
      "Outsigrid\n",
      "Aaakshy\n",
      "Chudesh\n",
      "Pattila\n",
      "Uhek\n",
      "Shkrita\n",
      "Alos\n",
      "Hivinendra\n",
      "\n",
      "\n",
      "Iteration: 300000, Loss: 15.246812\n",
      "\n",
      "Utiv\n",
      "Aorojamchhri\n",
      "Hreejk\n",
      "Hepajigyin\n",
      "Arshi\n",
      "Reeeeljsh\n",
      "Gyanund\n",
      "Againa\n",
      "Debaj\n",
      "Amala\n",
      "\n",
      "\n",
      "Iteration: 302000, Loss: 15.138795\n",
      "\n",
      "Geehup\n",
      "Sadeev\n",
      "Ripon\n",
      "Sndey\n",
      "Rirtet\n",
      "Arpesh\n",
      "Rbishas\n",
      "Jyarna\n",
      "Ahkraghanana\n",
      "Jeechat\n",
      "\n",
      "\n",
      "Iteration: 304000, Loss: 15.283723\n",
      "\n",
      "Nilant\n",
      "Rangor\n",
      "Rgumeshwor\n",
      "Echhangha\n",
      "Munind\n",
      "Mailambar\n",
      "Gabish\n",
      "Prisha\n",
      "Uevanand\n",
      "Bibachhar\n",
      "\n",
      "\n",
      "Iteration: 306000, Loss: 15.080416\n",
      "\n",
      "Yavil\n",
      "Sudila\n",
      "Janrishyo\n",
      "Bryeida\n",
      "Diphami\n",
      "Aie\n",
      "Shyatiran\n",
      "Ramevakant\n",
      "Parat\n",
      "Mankar\n",
      "\n",
      "\n",
      "Iteration: 308000, Loss: 15.226642\n",
      "\n",
      "Hadurak\n",
      "Maritu\n",
      "Rulpi\n",
      "Apeepjat\n",
      "Mist\n",
      "Jagoshshg\n",
      "Aspen\n",
      "Ramchanka\n",
      "Bivel\n",
      "Shameemo\n",
      "\n",
      "\n",
      "Iteration: 310000, Loss: 15.078121\n",
      "\n",
      "Canja\n",
      "Semidra\n",
      "Binan\n",
      "Uram\n",
      "Ushik\n",
      "Maresh\n",
      "Sonsan\n",
      "Ushemh\n",
      "Girlik\n",
      "Shttim\n",
      "\n",
      "\n",
      "Iteration: 312000, Loss: 15.266882\n",
      "\n",
      "Bramen\n",
      "Taimeshwral\n",
      "Lonal\n",
      "Yujau\n",
      "Ayojan\n",
      "Sukaram\n",
      "Usbichhendy\n",
      "Gubaum\n",
      "Boip\n",
      "Uhboran\n",
      "\n",
      "\n",
      "Iteration: 314000, Loss: 15.067550\n",
      "\n",
      "Bulkum\n",
      "Sunac\n",
      "Imw\n",
      "Aubash\n",
      "Sumala\n",
      "Runa\n",
      "Sumar\n",
      "Rubur\n",
      "Bangem\n",
      "Sugetant\n",
      "\n",
      "\n",
      "Iteration: 316000, Loss: 15.222138\n",
      "\n",
      "Rojeb\n",
      "Yasha\n",
      "Aawsh\n",
      "Ushin\n",
      "Umbik\n",
      "Aisni\n",
      "Una\n",
      "Jayeg\n",
      "Miichhi\n",
      "Ragini\n",
      "\n",
      "\n",
      "Iteration: 318000, Loss: 15.094516\n",
      "\n",
      "Eena\n",
      "Rshadhik\n",
      "Byeetra\n",
      "Ulvela\n",
      "Mesheka\n",
      "Uheshposh\n",
      "Aumja\n",
      "Sushan\n",
      "Ijesh\n",
      "Avit\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "file_name = \"combined_names.txt\"\n",
    "model_loss, generated_name = model(data,file_name,idx_to_char, char_to_idx, num_iter=320000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "num_iteration = list(range(0, len(model_loss)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x112fc4950>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.xlabel(\"Iterations in thousands\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.scatter(num_iteration,model_loss)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "160"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(model_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "generated_names_df=pd.DataFrame(generated_name, columns=[\"Generated Names\"]) \n",
    "generated_names_df = generated_names_df.replace('\\n','', regex=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "generated_names_df.to_csv('generated_names')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}